MareNostrum V is 8,000 nodes sitting in a Barcelona chapel, and getting code to run on it means SLURM time slots, airgapped storage and Amdahl’s Law. Here is what that setup teaches about scaling anything smaller.
Vector databases have become the default answer for AI agent memory, and for a project-scoped agent they are usually the wrong one. Plain Markdown files plus a local SQLite index give you grep, git diff and no server to babysit.
LLM agent memory architecture is what decides whether an agent holds up over a long task. This covers the four temporal scopes, the manage step almost everyone skips in the write-manage-read loop, and how summarization drift creeps in.
Standard vector RAG shreds documents into contextless fragments. Proxy-Pointer RAG keeps the heading hierarchy through skeleton trees and breadcrumbs, and on 66 queries against 10-K filings every number matched the ground truth.
Six things I ran into building a GPT-2 clone from scratch in PyTorch: the scaling flaw RsLoRA fixes, why RoPE replaced additive positional embeddings, why weight tying stopped paying off, and what KV-caching costs you in VRAM.
Deep Evidential Regression has the network predict Normal Inverse-Gamma parameters, so one forward pass separates noise in the data from what the model never learned. The awkward part is tuning the regularization term.
Linear regression handles more WordPress and WooCommerce work than the plugin marketing suggests. The math, RMSE against MAE, PHP inference without a Python bridge, and where regularization saves you.
Notes from WordCamp Asia 2026 in Mumbai: plugins and themes landed on every paid WordPress.com plan, the speed build put AI against raw FSE work, and Nirav Mehta’s session covered where LLMs still break. Here is what held up as useful for real builds.
WooCommerce ran a bug blitz: support engineers, not the core team, shipped more than 170 fixes across 20-plus products in a few weeks, helped along by Claude Code and a purpose-built dev environment skill. Here is how that workflow held together.
Reinforcement learning in Unity treated as state management rather than magic: the Bellman equation, Q-values for state and action pairs, the learning rate, and the epsilon-greedy trick that stops an agent from settling for a good enough path.